import type { ToolWithHandler } from '~/interfaces/MCPServer.js';
import { MCPOutputAdapter } from '~/utils/MCPOutputAdapter.js';
const outputAdapter = new MCPOutputAdapter();
export const recallTool: ToolWithHandler = {
name: 'recall',
description: `Retrieve memories from a role's semantic network
## Workflow
1. **DMN scan** — \`recall(role, null)\` → see the full memory landscape (hub nodes)
2. **Drill down** — pick keywords from the network → \`recall(role, "keyword")\` → get details
3. **Repeat** — follow new keywords in each result until you have enough context
> Always start with DMN (null query) to see what exists. Never guess keywords.
## Query Parameter
- \`null\` → **DMN mode** (recommended entry point): activates hub nodes, shows full network
- Single keyword: \`"PromptX"\` → spread from that node
- Multiple keywords: \`"PromptX testing fix"\` → multi-center activation
## Mode Parameter
- \`balanced\` (default): balance precision and association
- \`focused\`: precise lookup, frequent memories first
- \`creative\`: broad association, distant connections
## Examples
**DMN full scan:**
\`\`\`json
{ "role": "luban", "query": null }
\`\`\`
**Keyword drill-down:**
\`\`\`json
{ "role": "luban", "query": "PromptX testing", "mode": "focused" }
\`\`\``,
inputSchema: {
type: 'object',
properties: {
role: {
type: 'string',
description: 'Role ID to recall memories from, e.g.: java-developer, product-manager'
},
query: {
oneOf: [
{ type: 'string' },
{ type: 'null' }
],
description: 'Query keywords: string (space-separated) or null (DMN mode — auto-select hub nodes). Keywords must exist in the memory network.'
},
mode: {
type: 'string',
enum: ['creative', 'balanced', 'focused'],
description: 'Activation mode: creative (broad association), balanced (default), focused (precise lookup)'
}
},
required: ['role']
},
handler: async (args: { role: string; query?: string | null; mode?: string }) => {
const core = await import('@promptx/core');
const coreExports = core.default || core;
const cli = (coreExports as any).cli || (coreExports as any).pouch?.cli;
if (!cli || !cli.execute) {
throw new Error('CLI not available in @promptx/core');
}
// 构建 CLI 参数，支持 string | string[] | null
const cliArgs: any[] = [{
role: args.role,
query: args.query ?? null, // undefined转为null
mode: args.mode
}];
const result = await cli.execute('recall', cliArgs);
return outputAdapter.convertToMCPFormat(result);
}
};
